INDEX

Explanations

zip and zipper related words

The neuron strongly activates on tokens containing “zip” (e.g. zip, zipper, zipped, zippered, unzipping), i.e. references to a zipper or the act of zipping.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 イラストレーション

-1.73

 BRID

-1.66

LUD

-1.59

 VIDEOS

-1.57

 ORGANIC

-1.52

 名前

-1.50

 Trunks

-1.50

 DICK

-1.50

en

-1.48

 GREG

-1.48

POSITIVE LOGITS

1.70

Is

1.65

 είναι

1.62

archiwizowane

1.59

 sosteniendo

1.58

员

1.58

 Child

1.57

You

1.54

 Когда

1.54

 Community

1.53

Activations Density 0.010%